[cups.development] [RFE] STR #2762: handle error states for error log (i.e. do not repeat the same message again and again)

[Ulrich.Windl.rz.uni-regensburg]

DO NOT REPLY TO THIS MESSAGE.  INSTEAD, POST ANY RESPONSES TO THE LINK BELOW.

[STR New]

Network printers, when being used in a chaotic way (i.e. multiple users
print directly to the network port), may cause CUPS to have to wait for
the connection to become available. In such a case, the error log logs
many messages like:

E [27/Dec/2007:09:53:42 +0100] [Job 30] Unable to connect to printer
(retrying in 30 seconds): Connection timed out
E [27/Dec/2007:09:57:21 +0100] [Job 30] Unable to connect to printer
(retrying in 30 seconds): Connection timed out

(It would be useful to see to WHICH printer a connection timed out)

In stead of repeating the (nearly) same message over and over, one could
print syslog-like "last message repeated ## times", or, more general,
implement a state mechanism like this:

The printer connection state changed from "OK" to "ERROR". Log the change
of state, but do not log while the connection is in ERROR state. Once the
connection is OK again, log in (optionally). If the connection remains in
ERROR state for a longer time, use an exponentially increasing delay
between notifications (optional).

Maybe implement some simple event handling framework for that.

Link: http://www.cups.org/str.php?L2762
Version:  -feature